Our books

Become a Fan

« New Contributor | Main | Will great teaching land you a TT job? »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Marcus Arvan

Hi Mark: just gave your paper a quick read (I wish I had more time!), and have one big question. I will follow up on more later.

You suggest on pp. 1-2 that friend is a two-way relation -- that if no one is a friend to you, then you cannot be their friend. Moreover, you say there must be a shared understanding of friendship.

I expect many will find this plausible, but I'm not convinced myself -- in part because I think most people have an impoverished notion of friendship.

I consider myself a genuine *friend* to my students (well, at least the ones who aren't complete jerks!), even if they don't consider themselves *my* friend. Why? Because I care about them as individuals, and want the best for them.

One might reply that this is not real friendship -- that we should put the phrase "friend to my students" in scare-quotes. But I don't know. It all depends on what you think friendship is! I myself understand Aristotle as thinking that friendship is a genuine concern for the well-being of others as though they are part of *yourself*. He puts it, "Two souls in one body."

Well, can't I be a *friend* to you in this regard even if you don't conceive of yourself in this way, and even if you don't have the virtue of friendship to *me*? In short: why think that it has to be a two-way relation? Why not think that one can genuinely be a friend to those who are not friends to you?

Anyway, I'm sure many will be on board with how you're thinking about friendship -- and so perhaps you don't have to engage with these questions/worries of mine in the paper. But still, I'm curious what you think.

Mark Alfano

Thanks for your thoughts, Marcus.

A few thoughts in return:

1) I should probably make clearer in the paper that I'm not trying to give an interpretation of Aristotle. As far as I can tell, Aristotle thought that friendship was the only virtue that's relational in the way I describe. He seems to think of other moral virtues as monadic, in much the same way that neo-Aristotelians do.

2) Your claim that friendship is much broader than I admit is actually Aristotelian in spirit. He claims that bonds of friendship connect parent and child, host and guest, fellow travelers, even seller and buyer. But he also has a hierarchy of friendship, in which these examples are at the bottom of the list. Full-bore virtue friendships are, according to Aristotle, the sort of thing that you can only have with a few people. This is in part because of the intimacy they entail, in part because the amount of time it takes to know someone well to have a virtue friendship with them, and in part because the partiality of friendship means that friends can end up "competing" with each other for your time, attention, and resources.

3) You're right that friendship, for Aristotle involves "genuine concern for the well-being" of another person. I'm not sure I'd say it's concern for that person as if they were part of oneself, but that may be one way to read the "friend is another self" tagline. I'm pretty sure the line you mean to refer to is actually "one soul in two bodies," not "two souls in one body," by the way.

4) On to your objection on behalf of one-sided friendships... there are two things going on here. First, there's the question whether Aristotle thinks that there can be one-sided friendships; then there's the second question whether there really can be one-sided friendships. I'm quite sure that the answer to the first question is negative. In NE 1155-6, for example, Aristotle says that friendship involves not just reciprocation of goodwill but recognition or awareness of reciprocation of goodwill. On the second question, in a way it's just a matter of how you want to use the words, but I would say that the distinction I draw in the paper between someone who is friendly and someone who is a friend tracks an important distinction. You can say that someone who is friendly but not a friend still has the virtue of friendship, if you like, but it's worth keeping in mind that there's another (I would say, more important) kind of friendship.

Marcus Arvan

Mark: I have a general concern about your friendship model (i.e. extended-character hypothesis), as applied to the virtues you discuss (trustworthiness, humility, etc.).

You want to say that trustworthiness is not just causally dependent on others' attitudes towards you (e.g. their considering you worthy of truth), but *constitutively* dependent.

The former claim seems to me plausible. Someone's *thinking* I'm trustworthy can increase my trustworthiness (by getting me to think I am worthy of trust, which might cause me to become more trustworthy).

The idea that others' attitudes (even partially) constitute a person's trustworthiness seems to me much less plausible, and for one general reason: it would seem to entail that a person could become *less* trustworthy simply on account of others trusting them less. But that sounds strange.

It sounds less strange to me, I guess, if you go contextualist about it: trustworthiness is partially constituted by the context of judgment. For instance, I may be trustworthy-qua-acquaintance (e.g. you can trust me to be at the pub every night) but not trustworthy-qua-confidante (e.g. you can't trust me to keep a secret).

Is this something like what you have in mind -- that third-parties partially constitute virtues by establishing some kind of baseline context?

Mark Alfano

Marcus: What, other than strangeness, is the problem with one person's becoming less trustworthy just on account of others trusting them less? I'm failing to see the force of the objection. I don't have in mind the contextualist point, though I do think that most virtues have a built-in threshold, which allows for a decent amount of diversity in the dispositional signatures that can count as a given virtue. For instance, someone who is extremely trustworthy-qua-acquaintance and mostly trustworthy-qua-confidante and ... might count as trustworthy, whereas someone else who is mostly trustworthy-qua-acquaintance and completely trustworthy-qua-confidante and... might also count as trustworthy.

The point I'm trying to make in the paper, though, is that if someone's trustworthiness is functionally integrated with the trustingness of others (and he knows it and they know he knows it), then it might make sense to say that their dispositions and attitudes partially constitute his trustworthiness. Lots of things sound strange but are true. ;)

Marcus Arvan

Hi Mark: Thanks for the clarification. I guess I'm skeptical of the idea that a person's trustworthiness is functionally integrated with the trustingness of others.

Suppose Jones is a member of a very trusting community and Smith is a member of a very distrustful community, but that Jones and Smith are perfectly alike on the inside, as well as in terms of their dispositions (Jones keeps secrets just as well as Smith, etc.)

On your account, it would seem that Jones is more trustworthy than Smith just on account of how, in terms of integration with those around him, Jones is *considered* trustworthy whereas, say, Smith isn't.

This not only seems strange to me. It seems false. It seems to me that the right thing to say is:

(A) Jones and Smith are equally trustworthy *simply* because they are exactly the same on the inside.


(B) Different people and societies might have different relativized or contextualized *standards* of trustworthiness (i.e. you are trustworthy-to-me).

I'm fine with saying there are relativized or contextualized senses in which others' relations to me define (e.g.) my trustworthiness-to-them.

What I don't yet see is the case for saying that my trustworthiness *simpliciter* has anything to do with how others view me.

I could be *objectively* trustworthy, regardless of how others respond, what their standards are, etc.

Mark Alfano

Hi Marcus:

I'm willing to suppose that Jones and Smith are perfectly alike on the inside. I see no reason why I should grant that, given the very different communities that they're in, they are dispositional twins. That doesn't sound plausible at all, if they are the same internally.

If they were different internally, then I would be happy to grant that, in the face of different social/moral environments, they could be dispositional twins, but that's neither here nor there.

Marcus Arvan

Mark: if they are perfectly alike on the inside, how do they not have identical dispositional properties? If you swapped them from their respective cultures -- plopping Smith down in Jones' culture, and vice versa -- they would be disposed (thanks to their perfectly similar insides) to behave *just* as the other did in that culture, no?

So, it seems to me, a person's insides by themselves comprise a person's dispositional characteristics -- i.e. their inherent trustworthiness. Whether a given person or culture *considers* that person trustworthy (i.e. trustworthy-to-them) is another question.

Look, here's a broader way of putting the point. It seems to me that people outside of me could be systematically wrong about how trustworthy I am. Their standards for trustworthiness could be (morally) *incorrect*. I *am* trustworthy, their views on the matter be damned.

Anyway, I'm more than happy to accept your argument, but only if we're talking about a relativized form of the virtue: trustworthiness-for-*them*.

I think it makes perfect sense to say that trustworthiness-in-America is different than trustworthiness-in-Belize. But I still want to say that there is a broader, more objective, notion of trustworthiness that is inherent to a person (on the inside alone).

I know you want to deny this -- but what exactly is the argument? Your argument in the paper seems to me to at most establish that *some* construals of a virtue (i.e. trustworthiness-as-people-in-a-society-conceive-it) are partially comprised by external things. Why should one suppose that all are (that there are no purely internal notions of trustworthiness)?

Mark Alfano

Thanks for pressing the point. I think I'm starting to see where we really disagree.

If you rip me out of my social context, you change what's going on inside me too. Why? Some of my attitudes are about my social context, specifically about what other people think of me, expect of me, hope from me, and trust me to do/think/feel. If you rip me out of my social context and replace the contents of those attitudes with the relevant attitudes towards the people in the new context, I won't be the same inside. Additionally, if I am systematically interacting with a trusting person, which gives me the confidence, resolve, and self-knowledge to continue being trustworthy, then I am functionally integrated with that trusting person in exactly the way that Pritchard describes. Swap that trusting person for an untrusting person and all of a sudden I'm functionally integrated in a quite different way.

Now, on the question of the instrinsicness of dispositional properties: you say that virtues are on the inside, i.e., that they're intrinsic. I don't see why that's so. We know from the metaphysics literature that dispositional properties needn't be intrinsic. Examples include weight, vulnerability, recognizability, and vulnerability. (McKitrick, J., 2003a, ‘A Case for Extrinsic Dispositions’, Australasian Journal of Philosophy 81: 155–174.) Why think that virtues must be intrinsic, if other dispositions needn't be? Is there something special about virtues that forces them to be intrinsic?

The argument for thinking that virtues -- at least some of them -- are extrinsic is that we get more predictive and explanatory power, along with greater empirical adequacy, if we think of the relevant virtues in this way. It's not a deductive argument, but an inference to the best explanation. In the paper, I only try to establish this for friendship and for trustworthiness/trustingness. The final section says what it would take to make the case for generosity/gratitude and for humility/respect, but I don't attempt to do so in the paper. You may be right that the friendship model isn't a model of all virtues, but I'm not willing to admit that until specific cases have been carefully explored.

As to whether there's are "purely internal notions" of various virtues: I agree with you that there are such notions. Aristotle has such a notion. Contemporary neo-Aristotelians do too. For all I know, so do the folk. The question is whether anything answers to these notions. I'm more certain that something answers to the notion of trustworthiness I describe in the paper than that anything answers to the traditional, monadic notion.

Marcus Arvan

Hi Mark: cool, and thanks for such a clear and well thought out response. It's clear that you've thought through this issue very well. Not being all that familiar with the literature on dispositions, I don't feel well equipped to keep pressing you on it -- but in any case I find your reply pretty persuasive! :)

Mark Alfano

Thanks, Marcus. This has been a really helpful discussion for me, and will certainly make the paper better when I revise.

David Morrow

Hi Mark: I'm still working through your paper. I do hope to have comments for you soon.

Michael J. Augustin


I will begin looking at your paper tonight, with a view to sending you comments on it by tomorrow evening.

Anthony Carreras

Hi Mark,

Very cool and thought provoking paper! Here are some thoughts I have.

1) This is just a minor point, but I thought it worth mentioning that Aristotle himself acknowledges the distinction between someone who is friendly and someone who is a friend. The nameless virtue that he discusses in Book IV chapter 6 is something like "friendliness".

2) As you see it, what exactly are you gaining in this paper by appealing to friendship? The way I see it, you can make a compelling case that trustworthiness and some other virtues are relational in all the ways you describe without appealing to friendship (other than to note that friendship embodies and makes salient the correct model). That case would hinge on the strength of the situationist challenge and on the extended character thesis. How does *the fact that friendship is relational* add to this case? Here's another way of putting it: Are you appealing to friendship just because it provides a useful model? Or do you think that the fact that *friendship* provides this model lends support to the thesis that the virtues are relational?

I guess I'm trying to gauge the extent to which your paper endorses the following chain of reasoning: "Friendship is a virtue, a really important one, and it's relational. That alone provides some reason for thinking that the other virtues might be similarly relational." I'm skeptical of this reasoning because I think that while *being a friend* is a relational property, *being a good friend* is not (or, at least, is in some important ways not). So, I agree that my being your friend is constitutively dependent your being my friend, but I do not think that my being a good friend to you is constitutively dependent on your being a good friend to me. Many friendships are lopsided in this way, where A tends to go out of her way for B to an extent that B does not quite reciprocate to A. What this suggests to me is that what makes friendship a *virtue* (if it is a virtue) is not fully captured by the having of the de re attitudes (wish well to the other for the other's sake, in virtue of her character, etc) and the mutual awareness of those attitudes.

Mark Alfano

Thanks, Anthony.

ad 1): Yes, that's a good point. He also says something along those lines when discussing friendship as such.

ad 2): This is an important question. I think there are several things to say in response.

First, I think that friendship is worth exploring on its own because it seems, at first blush, to be unusual and important, and because it has not been sufficiently explored in the otherwise vast literature on Aristotle's ethics. It's unusual because it's so clearly relational. It's important because, at least according to Aristotle, it makes justice unnecessary, is a sort of pinnacle of virtue, and merits more of his attention than any other virtue.

Second, thinking through the nature of friendship can lead us to notice things about it that we then notice in other virtues. It's a recognized fact that it's easier to notice something if you're looking for it (though of course it's also easier to mistakenly think you've noticed something if you're looking for it, too). The friendship model prompts us to look for the interesting features of friendship in other virtues. I claim to have found some in the case of trustworthiness, though perhaps I'm just seeing things that aren't there. In a way, you might say that friendship really just provides a useful model, but I think that it may be necessary to use that model to grasp the structure of other virtues.

Third, another reason to explore the friendship model is that, if it can be established that at least one virtue is relational, then it becomes less eye-popping to claim that others are. I can imagine (hell, I can remember) someone saying, in response to my work on factitious virtue, "That's not really a virtue because it's relational." Neo-Aristotelians in particular are inclined to make such an argument. But they also tend to treat the Nicomachean Ethics as holy writ, so if I can show them that they themselves are committed to there being at least one relational virtue, my argumentative task is easier.

Fourth, I'm not sure how to respond to your distinction between being a friend and being a good friend. In particular, I take it that being a good friend entails being a friend (indeed, is partially constituted by being a friend). But then, presumably that means that being a good friend is relational too. Perhaps what you have in mind is this: one of the things I claim about friendship is that friends come in pairs; for X to be a friend, there must be a Y distinct from X such that Y is also a friend (and is, moreover, a friend of X). But, you say, good friends don't necessarily come in pairs: for X to be a good friend, there needn't be a Y distinct from X such that Y is also a good friend (of X). Instead, all that needs to hold is that there be a Y distinct from X who is a friend (good, bad, or neutral). That's an interesting point, which seems to me correct. How much does it undermine the friendship model? Well, if you think that only good friendship is a virtue, it might undermine the model. I guess I think that even middling friendships are virtues, and I'm inclined to say that bad friend isn't really a friend at all. So maybe the distance between our views isn't that great.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)

Open job-market threads